INTRODUCTION — Chronic pain is one of the most common reasons that patients seek medical attention. Chronic pain results from combined biologic, psychologic, and social factors, and most often requires a multifactorial approach to management. In addition to nonpharmacologic therapies, many patients require medications to manage pain. This topic will discuss an approach to pharmacologic management based on the type of pain, and an overview of drug choices.

The general approach to management of chronic non-cancer pain and nonpharmacologic therapies are discussed separately. Evaluation of chronic pain, and the use of opioids for chronic non-cancer pain are also discussed separately.

●(See "Approach to the management of chronic non-cancer pain in adults".)

●(See "Evaluation of chronic non-cancer pain in adults".)

●(See "Use of opioids in the management of chronic non-cancer pain".)

GENERAL PRINCIPLES OF DRUG SELECTION — The optimal choice of pharmacologic therapy depends on the type of chronic pain syndrome. In particular, neuropathic pain should be distinguished from nociceptive pain since treatments differ. For patients who are thought to have predominantly nociceptive pain, if usual pharmacological treatment (as part of a multidisciplinary treatment plan) proves ineffective, reconsider whether the patient may have neuropathic or centralized pain, or a combination of both, and modify treatment accordingly (table 1 and algorithm 1).

The patient's medical status (eg, cardiovascular, hepatic, renal, and cognitive issues) may affect the choice of drug, due to potential for drug side-effects, drug clearance, and drug-drug interactions. Consistent with a biopsychosocial model of care, multimodal therapy that includes alleviation of disturbed sleep or depressed or anxious mood can significantly increase response to other drug and interventional treatments [1]. Involving family and caregivers may improve prescription compliance, and knowledge about social support (eg, assistance with transportation) may affect decisions about the frequency of follow-up visits for monitoring and adjustment of drug doses.

We always reserve opioids for patients in whom other therapies are either ineffective or contraindicated. We prescribe opioids at lowest dose and for the shortest intervals possible, and only if they measurably improve function and quality of life. (See 'General considerations' below.)

Pain pathways and the effects of drugs on those pathways are shown in a figure (figure 1). Mechanisms of analgesic drug classes are shown in a table (table 2).

Pharmacologic therapy for nociceptive pain — Nonsteroidal antiinflammatory drugs (NSAIDs) are the mainstay of treatment of musculoskeletal pain if pharmacologic therapy is required in addition to multimodal therapy. However, the efficacy of NSAIDs for chronic musculoskeletal pain without evidence of ongoing inflammation is low (eg, chronic low back pain) [2]. Whenever possible, structural, inflammatory, or disease-related causes for pain should be targeted, which may reduce or eliminate the need for analgesics. (See "Approach to the management of chronic non-cancer pain in adults", section on 'Start with nonpharmacologic therapy'.)

For patients with musculoskeletal pain with multiple joints involved, and without contraindications to NSAIDs, oral NSAIDs are appropriate first-line medications. Topical NSAIDs may be used for patients with localized arthritis of superficially located joints (eg, hand and wrist, foot and ankle, or knee), with the potential to achieve therapeutic synovial tissue drug levels with lower systemic drug levels compared with oral therapy. Topical agents may also be appropriate for focal myofascial pain conditions. (See 'Topical agents' below.)

Acetaminophen is not considered a first-line therapy for chronic osteoarthritis or back pain, due to lack of efficacy and concerns for hepatotoxicity when used at a higher than recommended dose. (See 'Acetaminophen' below.)

For patients who are presumed to have nociceptive pain, when NSAIDs prove ineffective, the diagnosis of central sensitization or neuropathic pain, or a combination of both could be considered, and a pain-relieving antidepressant or possibly antiepileptic be used as an alternative. (See 'Pharmacologic therapy for neuropathic pain' below.)

We avoid using opioids long-term whenever possible in patients with nociceptive pain, and use them only when the benefits outweigh the potential risks, when other therapies have failed to provide adequate pain relief and improvement in function, at the lowest effective dose and for shortest duration, and usually combined with other nonopioid and nonpharmacologic therapy. While opioids can be highly effective for acute pain, the role of opioids for chronic nociceptive pain is uncertain (see "Prescription of opioids for acute pain in opioid naïve patients"). In a 2018 meta-analysis of randomized trials of patients with chronic non-cancer pain, the benefits of opioids with respect to pain and function were similar to the benefits of nonopioid alternatives, though this was based on low to moderate quality studies [3]. (See 'General considerations' below and "Use of opioids in the management of chronic non-cancer pain", section on 'Indications for opioid therapy'.)

Pharmacologic therapy for osteoarthritis and chronic back pain are discussed in depth separately. (See "Overview of the management of osteoarthritis", section on 'Pharmacologic therapy' and "Subacute and chronic low back pain: Nonpharmacologic and pharmacologic treatment", section on 'Adjunctive pharmacologic therapy for patients with persistent, significant symptoms'.)

Pharmacologic therapy for neuropathic pain — For most patients with chronic neuropathic pain, initial treatment involves either selected antidepressants (ie, tricyclic antidepressants [TCAs] or serotonin-norepinephrine reuptake inhibitors [SNRIs]), or antiepileptic medications (ie, gabapentin or pregabalin), with adjunctive topical therapy (eg, topical lidocaine, 8% capsaicin patch) when pain is localized (table 3) [4-8].

The effects of systemic medications on receptors and neurotransmitters involved with pain are shown in a table (table 4). (See 'Antidepressants' below and 'Antiepileptic medications' below.)

The choice among treatments should be individualized based on the pain condition (if known), patient-specific characteristics, comorbid conditions, medication side effect profile, cost, and patient values and preferences. In practice, comorbidities and concurrent medications often favor one drug class or another (eg, start with an antidepressant if the patient is also depressed or anxious, or a gabapentinoid when antidepressant drug-interactions or side-effects are problematic) (table 5). The narrower choice within a class is also influenced by the patient's age, preferences with regard to dosing frequency, side effects, and cost/formulary considerations.

Combination therapy with both antidepressant and antiepileptic medications is often required, because less than one-half of patients with neuropathic pain will respond to a single agent [9]. However, evidence is scant regarding the efficacy of specific combinations and the safety of combination treatment [10,11].

Opioids should be considered a second- or third-line option, especially when there is an expectation that they may be prescribed long-term. Opioids may be considered earlier in the treatment of select patients, such as those with high severity intractable pain, episodic exacerbations of severe pain, or neuropathic cancer pain [7]. The efficacy of opioids for neuropathic pain is uncertain. In a 2013 review of 31 randomized trials of opioids for neuropathic pain, among patients treated for several days to 12 weeks, 57 percent of patients experienced 33 percent pain relief with opioids, compared with 34 percent of patients who received placebo, but risk of bias in the included studies was high (table 3) [8].

Choices of pharmacologic agents for specific neuropathic conditions are discussed separately in multiple topic reviews.

●(See "Postherpetic neuralgia", section on 'Treatment'.)

●(See "Management of diabetic neuropathy", section on 'Pain management'.)

●(See "Complex regional pain syndrome in adults: Treatment, prognosis, and prevention", section on 'Approach to pharmacotherapy'.)

●(See "Trigeminal neuralgia", section on 'Medical treatment'.)

●(See "Initial treatment of fibromyalgia in adults", section on 'Selection of further drug therapy'.)

●(See "Epidemiology, clinical manifestations, diagnosis, and treatment of HIV-associated peripheral neuropathy".)

NONOPIOID ANALGESICS — Nonsteroidal antiinflammatory drugs (NSAIDs), acetaminophen, norepinephrine re-uptake inhibiting antidepressants, and anticonvulsants are the most commonly used nonopioid analgesics (table 6).

Nonsteroidal antiinflammatory drugs — Both the nonselective NSAIDs and the more selective cyclooxygenase 2 (COX-2) inhibitors are primarily indicated for mild to moderate pain, particularly of musculoskeletal origin, although a number of newer compounds carry an indication for severe pain. NSAIDs may be useful for chronic pain that persists due to underlying inflammatory mechanisms (eg, arthritis). Usual doses, selected characteristics, and therapeutic roles of NSAIDs are shown in a table (table 6).

The primary effect of NSAIDs is to inhibit cyclooxygenase (COX). COX catalyzes arachidonic acid metabolism yielding prostanoids, which play a significant role in inflammation and certain pain conditions. The analgesic effect of NSAIDs derives both from its peripheral action on the enzyme COX, as well as an effect on the central nervous system, at least in certain pain states [12]. Inflammation sensitizes peripheral nociceptors, which may create a hyper-nociceptive milieu and thereby increase ascending nociceptive signaling [13]. (See "NSAIDs: Pharmacology and mechanism of action" and "NSAIDs: Therapeutic use and variability of response in adults" and "Overview of COX-2 selective NSAIDs".)

NSAIDs are synergistic with opioids, producing a modest dose-sparing effect in postoperative pain, though not studied in chronic pain. The authors will trial a combination of NSAIDs and opioids when inflammatory nociceptive pain is concurrent, such as rheumatoid arthritis. There is clear individual variation in therapeutic and adverse responses to NSAIDs. NSAIDs are associated with gastropathy, renal toxicity, platelet inhibition, and cardiovascular risk, in addition to drug-drug interactions. For specific interactions, use the Lexicomp drug interactions program included with UpToDate.

For patients who chronically take NSAIDs, treatment trials on and off the NSAIDs should be considered, to avoid unnecessary polypharmacy and attendant risks.

Therapeutic uses for NSAIDs, including selective COX-2 inhibitors, and adverse effects are discussed separately. (See "NSAIDs: Therapeutic use and variability of response in adults" and "Nonselective NSAIDs: Overview of adverse effects" and "NSAIDs: Adverse cardiovascular effects" and "Overview of COX-2 selective NSAIDs".)

Acetaminophen — Though commonly used, evidence of efficacy of acetaminophen for chronic pain is very limited, with at best a subset of patients receiving sustained meaningful analgesia [14,15]. The analgesic mechanism of acetaminophen is uncertain [16]. Acetaminophen does not have peripheral anti-inflammatory effects, but is sometimes categorized as an NSAID, since evidence suggests a predominantly central nervous system effect on prostaglandin synthesis [17]. Despite the sparse evidence, acetaminophen provides analgesia for some patients. Thus, it is reasonable to consider acetaminophen as an adjunct for mild to moderate musculoskeletal pain or episodes of acute or chronic pain, in patients who report benefit.

The combination of acetaminophen with an NSAID has not been evaluated for chronic pain, though this combination is commonly used for acute pain.

There is controversy over the maximum safe daily dose of acetaminophen, primarily related to liver toxicity. The US Food and Drug Administration (FDA) recommended maximum dose of acetaminophen is 4 g per day [18]. However, for long-term use, many experts and manufacturers limit the daily dose to 3000 mg orally (2000 mg per day for older patients and for patients with liver disease). (See "Acetaminophen (paracetamol): Drug information" and "Acetaminophen (paracetamol) poisoning in adults: Pathophysiology, presentation, and evaluation".)

In addition to hepatotoxicity, chronic acetaminophen use may be associated with chronic kidney disease, hypertension, chronic daily headache, and peptic ulcer disease. (See "Epidemiology and pathogenesis of analgesic-related chronic kidney disease", section on 'Acetaminophen' and "NSAIDs and acetaminophen: Effects on blood pressure and hypertension", section on 'Effects of acetaminophen on blood pressure' and "Unusual causes of peptic ulcer disease", section on 'Non-NSAID medications'.)

ANTIDEPRESSANTS — Some specific categories of antidepressants are first-line treatments for many neuropathic pain disorders. Antidepressants are a heterogeneous group of medications approved to treat major depressive disorders. Both tricyclic antidepressants (TCAs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) possess analgesic qualities. The evidence for the effectiveness of highly selective serotonin reuptake inhibitors (SSRIs) is weaker, and completely lacking for musculoskeletal pain. SSRIs are not considered first-line treatment for any chronic pain condition [19]. (See "Tricyclic and tetracyclic drugs: Pharmacology, administration, and side effects" and "Serotonin-norepinephrine reuptake inhibitors (SNRIs): Pharmacology, administration, and side effects" and "Selective serotonin reuptake inhibitors: Pharmacology, administration, and side effects".)

TCAs or SNRIs are indicated for treatment of pain even in the absence of mood disturbance. Analgesic antidepressants provide pain relief separate from their antidepressant effects, with analgesic effects known to occur in non-depressed patients. However, for some patients, an effect on underlying depression, especially for SSRIs, may also contribute to relief of pain. (See "Evaluation of chronic non-cancer pain in adults", section on 'Psychiatric comorbidity'.)

Multiple studies have found that TCAs or SNRIs may provide effective pain relief in neuropathic pain conditions with a number needed to treat of 3.6 to 6.4, for one patient to achieve at least 50 percent pain relief [4]. Antidepressants may be effective for a wide range of pain conditions, including low back pain, various types of neuropathic pain, and fibromyalgia [20-24] (see "Treatment of fibromyalgia in adults not responsive to initial therapies"). Antidepressants with the greatest effect upon norepinephrine seem to have the greatest analgesic effect [20,21]. The analgesic response from norepinephrine re-uptake inhibition results from subsequent up-regulation of the descending inhibitory pain pathways [25,26].

For low back pain unresponsive to nonpharmacologic therapy the authors frequently prescribe antidepressants with analgesic properties and individualize the choice of drug and dosing. Back pain is often caused by a wide range of biomedical and psychosocial factors and is best treated with multimodal nonpharmacologic and pharmacologic therapies. The quality of evidence supporting the use of antidepressants for low back pain is generally low; efficacy of therapy for low back pain is difficult to study, and no medication or individual therapy has evidence for more than a modest beneficial effect. (See "Subacute and chronic low back pain: Nonpharmacologic and pharmacologic treatment".)

Noticeable analgesic (and antidepressant) effects may require two to four weeks of antidepressant therapy, possibly due to the time necessary to increase production of neuroprotective proteins, such as brain-derived neurotrophic factor [27]. In the authors' experience, many purported treatment failures are a result of too low a dose and/or too short a duration for a clinically meaningful improvement to occur. Doses of antidepressants used for chronic pain are shown in a table. (table 3)

Tapering and discontinuing antidepressants can be challenging. Abruptly stopping or rapidly tapering can cause a variety of symptoms, including agitation, anxiety, chills, diaphoresis, dizziness, dysphoria, fatigue, headache, insomnia, irritability, myalgias, nausea, paresthesias, rhinorrhea, and tremor. A very slow tapering schedule over two to four weeks is recommended, though the authors have at times added a low-moderate dose of the long-acting (LA) SSRI fluoxetine (eg, 10 to 20 mg) while discontinuing the initial antidepressant over two to four weeks, and then discontinuing the fluoxetine. (See "Discontinuing antidepressant medications in adults".)

Tricyclic antidepressants — TCAs are a mainstay of treatment for a variety of chronic pain states, with or without coexisting depression, despite the fact that none of the TCAs carry a US Food and Drug Administration (FDA) labeled primary indication for pain management. Most of the studies supporting their use are smaller and older than studies of SNRIs.

TCAs can be divided into tertiary amines and their demethylated secondary amine derivatives. In addition, maprotiline (Ludiomil) is often considered in this drug class, although it is a tetracyclic antidepressant. Amitriptyline has been the most widely studied TCA in chronic pain [28,29]. A number of others, including doxepin, imipramine, nortriptyline, and desipramine also have been used effectively. Amitriptyline is the only TCA with proven efficacy for prevention of migraine, though others are widely used. (See "Preventive treatment of episodic migraine in adults", section on 'Antidepressants'.)

TCAs have independent analgesic effects in addition to relieving the depressive symptoms associated with chronic pain [30]. Most TCAs have antihistaminic side effects that may be desirable in patients who have problems with sleep initiation and maintenance. Anticholinergic effects do not appear to contribute to analgesia, are common and may lead to dose limitation and discontinuation.

●Dosing – TCAs should be started at a low dose, with dosing slowly escalated as tolerated. As an example, nortriptyline may be started at 10 mg/day orally, increasing the dose at weekly intervals in 10 to 25 mg increments based on response and tolerability to a maximum dose of 75 to 150 mg daily at bedtime. It can take up to 6 to 12 weeks, including two weeks at the highest dose tolerated, for an adequate trial of treatment with TCA. Although effective analgesia has been reported at lower doses than those required for treatment of depression [7,31], there is no strong evidence to support this approach. The authors do sometimes prescribe at higher antidepressant level doses when analgesic targets are unmet at lower doses and adverse effects are not dose limiting. The authors have observed that many purported treatment failures are a result of too low a dose and/or too short a duration for a clinically meaningful improvement to occur (table 3).

Starting doses of tricyclics should be reduced by one-half for older patients, who should be watched carefully for untoward effects as the doses are escalated slowly. Note that the American Geriatrics Society Beers Criteria for Potentially Inappropriate Medication Use in Older Adults includes TCAs [32].

●Side effects – TCAs are associated with multiple undesirable adverse effects that vary depending on the individual agent (table 7). Adverse effects are dose-related, and include anticholinergic effects, antihistaminic effects, alpha-1 adrenergic receptor blockade, and cardiac effects (ie, increasing intraventricular conduction, prolonged QT interval, prolonged conduction through the atrioventricular node). TCAs are relatively contraindicated in patients with severe cardiac disease, particularly conduction disturbances. Though historically recommended [7], the authors do not routinely obtain a pre- and post-treatment ECG unless there is a history of cardiac disease or arrhythmias, or if the TCA will be co-prescribed with antiarrhythmic drugs. (See "Tricyclic and tetracyclic drugs: Pharmacology, administration, and side effects", section on 'Side effects'.)

Among the TCAs, amitriptyline is the most sedating, and also has the most potent anticholinergic effects. Thus we rarely prescribe amitriptyline as the first choice TCA for chronic pain, unless sleep initiation and maintenance is a problematic complaint. Nortriptyline is our preferred initial TCA, since it is less sedating and with fewer anticholinergic side effects than amitriptyline. Anticholinergic adverse effects including dry mouth, orthostatic hypotension, constipation, and urinary retention can be reduced by starting with low doses administered at bedtime and with slow titration to higher dose.

Use of TCAs for specific chronic pain conditions is discussed separately. (See "Postherpetic neuralgia", section on 'Tricyclic antidepressants if gabapentinoids ineffective or not tolerated' and "Management of diabetic neuropathy", section on 'Tricyclic drugs' and "Complex regional pain syndrome in adults: Treatment, prognosis, and prevention", section on 'Adjuvant medications for neuropathic pain'.)

Serotonin norepinephrine reuptake inhibitors — Among the SNRIs, venlafaxine and duloxetine have been used for treatment of peripheral neuropathic pain, duloxetine and milnacipran have been used for treatment of fibromyalgia, and duloxetine has best evidence of efficacy for chronic musculoskeletal pain. Milnacipran is a more potent inhibitor of norepinephrine reuptake than the other SNRIs and would therefore be expected to be a more potent analgesic, but analgesic response has not been directly compared among drugs in this class. (See "Serotonin-norepinephrine reuptake inhibitors (SNRIs): Pharmacology, administration, and side effects".)

Duloxetine — Of all antidepressants, duloxetine has the largest evidence base to support analgesic efficacy, and is effective and FDA approved for the treatment of painful diabetic neuropathy, fibromyalgia, chronic low back pain, and osteoarthritis [33-35]. In addition to these indications, duloxetine is approved by the FDA for treatment of major depression, anxiety, and stress urinary incontinence.

The most common side effects include nausea, dry mouth, insomnia, drowsiness, constipation, fatigue, and dizziness (table 7). Side effects are reduced by administering duloxetine 30 mg orally once daily for one week before increasing to the usual dose of 60 mg once daily. The authors will occasionally increase the dose as high as 120 mg daily in order to maximize analgesia from the increased norepinephrine reuptake that occurs at higher dose, with attention to potential for serotonin side effects.

Duloxetine should be avoided in patients with hepatic or severe renal insufficiency. Gradual tapering is recommended at discontinuation to avoid withdrawal symptoms.

Doses and efficacy of duloxetine for painful diabetic neuropathy, chronic back pain, and fibromyalgia are discussed separately. (See "Management of diabetic neuropathy", section on 'Duloxetine' and "Subacute and chronic low back pain: Nonpharmacologic and pharmacologic treatment", section on 'Second-line pharmacologic therapy' and "Management of moderate to severe knee osteoarthritis", section on 'Duloxetine' and "Initial treatment of fibromyalgia in adults", section on 'Duloxetine'.)

Milnacipran — Milnacipran is a newer SNRI approved by the FDA for fibromyalgia, and in Europe and Japan as an antidepressant. It inhibits norepinephrine reuptake more so than serotonin inhibition, and hence may be more potent for the management of both neuropathic and centralized pain conditions. (See "Serotonin-norepinephrine reuptake inhibitors (SNRIs): Pharmacology, administration, and side effects", section on 'Milnacipran'.)

Venlafaxine — Venlafaxine may be used to treat acute and chronic neuropathic pain [36]. At very low doses venlafaxine has activity similar to an SSRI, but its norepinephrine reuptake properties prevail as the dose is increased (table 3).

Cardiac conduction abnormalities have been reported in a small number of patients, and blood pressure increases can occur; therefore, venlafaxine should be prescribed with caution in patients with cardiac disease (table 7). Discontinuation symptoms are particularly likely with discontinuation of venlafaxine due to its short half-life.

The use of venlafaxine for treatment of diabetic neuropathy is discussed separately. (See "Management of diabetic neuropathy", section on 'Venlafaxine'.)

ANTIEPILEPTIC MEDICATIONS — Three antiepileptic medications (gabapentin, pregabalin, and carbamazepine) are among the five drugs approved by the US Food and Drug Administration (FDA) for the treatment of neuropathic pain (table 3) [37].

Gabapentin and pregabalin — The gabapentinoids have proven efficacy versus placebo in several neuropathic pain conditions [7,38-40]. Gabapentin has primarily been studied and found effective for the treatment of postherpetic neuralgia and painful diabetic neuropathy; evidence for efficacy in other types of neuropathic pain is limited [41,42]. A 2019 systematic review of 45 randomized trials of pregabalin in patients with postherpetic neuralgia, painful diabetic neuropathy, or mixed neuropathic pain found pregabalin at doses of 300 mg to 600 mg daily more effective than placebo [43]. Response rates were lower for patients with central neuropathic pain or fibromyalgia, compared with patients with postherpetic neuralgia or painful diabetic neuropathy. There was no evidence of efficacy for patients with HIV neuropathy. Pregabalin is the only FDA approved medication for the treatment of neuropathic pain associated with spinal cord injury.

Gabapentin and pregabalin bind to the voltage-gated calcium channels at the alpha 2-delta subunit in the central nervous system. While the binding sites are well characterized, the mechanisms leading to pain relief are likely diverse [44,45].

●Gabapentin dosing – Treatment with gabapentin should be initiated at a low dose with gradual increases until pain relief or dose-limiting adverse effects are achieved. Gabapentin is typically initiated at 300 mg at night. In patients with sensitivities or vulnerability to nocebo effects, consider a starting dose of 100 mg at night to minimize initial side effects.

The typical effective daily dose range for immediate-release (IR) gabapentin is 1200 to 2400 mg/day on a three times a day schedule, with a maximum daily dose of 3600 mg. As gabapentin can be sedating, the authors at times dose it asymmetrically with a larger dose at night to facilitate sleep. For the extended release formulation, one regimen starts with 300 mg orally once daily, gradually increasing to a maximum of 1800 mg orally once daily if needed. Adjustment for renal impairment is required for both immediate and extended release, and use of extended-release (ER) is not recommended in patients with severe renal impairment.

Gabapentin may not be reliably absorbed due to an active L-amino transport system in the proximal small bowel that is both saturable and dose-dependent. An adequate trial of treatment with gabapentin can require two months or more [41].

●Pregabalin dosing – The recommended starting dose for pregabalin (IR) is 150 mg, divided into two or three doses daily, and increased to a total daily dose of 300 mg based upon tolerability and effect. Further titration (to 600 mg daily) after two to four weeks may be considered if necessary. As with gabapentin, the authors at times dose it asymmetrically with a larger dose at night to facilitate sleep. An extended-release preparation is FDA approved for once-daily dosing for diabetic peripheral neuropathy and post-herpetic neuralgia, but not for fibromyalgia. Adjustment for renal impairment is required for both immediate and extended release, and use of ER is not recommended in patients with severe renal impairment.

Pregabalin may provide analgesia more quickly than gabapentin, both because a lower initial dose may be efficacious and because a shorter time is required to titrate to a full dose [46]. Pregabalin is well absorbed with high bioavailability.

●Adverse effects – Gabapentin and pregabalin can produce dose-dependent dizziness and sedation that can be reduced by starting with lower doses and titrating slowly. Importantly, respiratory depression has been reported in older patients and in those who receive gabapentin along with other analgesics and sedatives [47,48]. In a population-based study of approximately 6500 patients who were receiving prescription opioids, concomitant prescription of pregabalin was associated with a dose-related increase in the risk of opioid related mortality (odds ratio [OR] 1.52 [95% CI 10.4-2.2] for low dose pregabalin, OR 2.0 [95% CI 1.24-5.06] for high dose pregabalin) [49]. Similar results have been reported for coadministration of opioids and gabapentin [50]. The data regarding both respiratory depression and abuse potential with both drugs are evolving.

Gabapentinoids may be associated with increased risks of mental health disturbances (eg, depression, suicide), unintentional overdose, and motor vehicle accidents [51,52]. Caution is also advised when prescribing these drugs to incarcerated patients and individuals with substance use disorders, as this class of medications has abuse potential due to their psychoactive effects, especially at higher doses [53,54]. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects".)

Other antiepileptics — Data informing the use of carbamazepine and oxcarbazepine for neuropathic pain are limited. Carbamazepine is the first-line therapy for trigeminal neuralgia, and is best studied for this indication. Oxcarbazepine is an alternative for patients who don't respond to or tolerate carbamazepine. (See "Trigeminal neuralgia", section on 'Medical treatment'.)

Some evidence from short-term trials suggests carbamazepine is moderately effective for other chronic neuropathic pain [55], but its use may be limited by side effects.

Other antiepileptic medications, including topiramate, lamotrigine, levetiracetam, phenytoin, sodium valproate, zonisamide, tiagabine, and the benzodiazepine clonazepam, have been utilized anecdotally and in randomized trials for various pain conditions [56]. Evidence of their effectiveness is limited, and in general these agents should be reserved for specialty care, third-line treatment in patients who don't respond to or cannot tolerate other medications. A 2013 systematic review of lamotrigine for acute and chronic pain concluded that it does not have a place in the treatment of pain, given other more effective therapies [57]. A 2014 systematic review of studies of levetiracetam found no evidence to support its use for neuropathic pain [58], consistent with a strong recommendation against use in a 2015 review of pharmacotherapy for neuropathic pain [4].

Laboratory monitoring for patients who receive anticonvulsants is discussed separately. (See "Initial treatment of epilepsy in adults", section on 'Laboratory monitoring'.)

OTHER ADJUVANT MEDICATIONS — When usual medication regimens are not effective, a degree of exploration and innovation may be required, including the administration of other adjuvant medications to potentiate analgesia. To minimize polypharmacy, on-off-on trials may help determine if medications are beneficial; they should be discontinued if improvement does not occur.

Other than laxatives, we rarely add drugs to treat side-effects associated with analgesics, and certainly do not recommend the addition of stimulants to overcome sedation.

Topical agents — Topical agents have several potential advantages over systemic drugs for pain: delivery at the painful site of injury, lower initial rates of systemic absorption and so fewer systemic effects, and patient preference. However, significant systemic concentrations and systemic side-effects are possible. Topical agents may be offered as first-line therapy for very well-localized nociceptive or neuropathic pain conditions, though are more commonly prescribed as an adjunct to systemic medication.

●Transdermal nonsteroidal antiinflammatory drugs – Topical nonsteroidal antiinflammatory drugs (NSAIDs), in the form of a gel, spray, or cream, can provide relief for acute musculoskeletal pain [59] and may be beneficial in patients with single joint osteoarthritis [60]. Evidence of effectiveness for these agents for other chronic pain conditions is lacking. The risk of gastrointestinal, renal, and cardiovascular toxicity is much lower with topical NSAIDs as compared with its oral formulation due to the reduced systemic absorption (5- to 17-fold lower for topical diclofenac compared with oral [61,62]). Topical NSAIDs may be better tolerated than oral preparations, with mild skin rashes being the most commonly reported side effect. (See "Management of knee osteoarthritis", section on 'Topical NSAIDs'.)

●Topical lidocaine – Topical lidocaine is considered second-line therapy for some forms of neuropathic pain [4]. For chronic pain, topical lidocaine is usually used in the form of a patch or plaster. Data supporting the efficacy of topical lidocaine are limited [63], and the best evidence suggests that it may be beneficial for postherpetic neuralgia, and possibly for painful diabetic neuropathy. (See "Management of diabetic neuropathy", section on 'Topical therapies or neuromodulation' and "Postherpetic neuralgia", section on 'Topical therapy for patients with milder symptoms'.)

A single 5% lidocaine patch contains 700 mg lidocaine. Up to three patches can be applied in a single application in a 24 hour period, for up to 12 hours, leaving a patch-free period of at least 12 hours. Systemic absorption of lidocaine when used according to this regimen is low (approximately three percent) [64], but topical lidocaine should be used cautiously in patients with hepatic, renal, or cardiac dysfunction. (See "Postherpetic neuralgia", section on 'Topical therapy for patients with milder symptoms'.)

●Capsaicin cream – Capsaicin cream has been used in patients with postherpetic neuralgia, HIV neuropathy, diabetic neuropathy, and in patients with osteoarthritis in one or a few joints. (See "Postherpetic neuralgia", section on 'Topical therapy for patients with milder symptoms' and "Overview of the management of osteoarthritis", section on 'Pharmacologic therapy' and "Management of knee osteoarthritis", section on 'Topical capsaicin' and "Epidemiology, clinical manifestations, diagnosis, and treatment of HIV-associated peripheral neuropathy".)

Capsaicin is an alkaloid derived from chili peppers; repeated application is thought to deplete substance P from primary afferent neurons [65]. It is available over-the-counter as a cream (0.025 or 0.1%), and as a high concentration patch (8%, which requires a prescription, and at this dose requires pre-treatment with topical or injectable lidocaine). Capsaicin cream must be applied three to four times per day over the entire painful area for up to six to eight weeks before optimal pain relief can be achieved, while the patch, administered under close clinician supervision, is applied as a single 60 minute application. The major adverse effects of capsaicin are burning, stinging, and erythema at the site of application, leading to intolerance in up to one-third of patients.

Cannabis and cannabinoids — The use of cannabis and cannabinoids for chronic pain is controversial, and is complicated by the mixed and varied legal status of these substances in various locations. The authors practice in a state where recreational and medicinal cannabis use is legal; considerations with respect to marijuana include the following:

●We recommend against marijuana in patients with cannabis use disorder or prior or current substance use disorder.

●Importantly, we recommend discontinuation of cannabis that has been used as an analgesic but has not reduced pain or improved function and quality of life.

●We warn patients about the impact of cannabis on safe automobile and other equipment operation, especially in combination with alcohol.

●We do not routinely discontinue prescription of other prescribed analgesic drugs, including opioids, when the patient reports medicinal or recreational cannabis use, as long as the patient is an active participant in all other recommended aspects of multimodal pain management.

●We no longer routinely monitor urine for the presence of cannabis, since its biological presence lasts weeks (when used episodically) or months (when used daily).

Systematic reviews and meta-analyses of trials including multiple patient populations and formulations of cannabis and cannabinoids have reported mixed results on efficacy for chronic pain [66-74]. A 2017 report from the National Academies of Sciences, Engineering and Medicine found evidence of significant reduction in chronic pain with the use of cannabis or cannabinoids. In a 2018 meta-analysis of 47 randomized trials of cannabis use for various types of chronic pain, there was moderate evidence that cannabis reduced pain by 30 percent, however, adverse event rates were high [66]. Based on pooled event rates, the number needed to treat for benefit was 24 (95% CI 15-61) and the number needed to harm was 6 (95% CI 5-8). In most studies, cannabis was used as an adjunct to other pain therapies.

In addition to cannabis, cannabinoid-type drugs are widely and increasingly commercially available in the United States. An increasing number of clinicians are living in states that have legalized medical cannabis, and it is likely that patients with chronic pain will seek approval or a specific recommendation for use of cannabis or cannabidiol (CBD). Available preparations and issues related to the use of these medications for chronic pain, including cancer pain, are discussed separately. (See "Cancer pain management: Role of adjuvant analgesics (coanalgesics)", section on 'Cannabis and cannabinoids'.)

Short-term adverse effects of cannabis and cannabinoids include dizziness, dry mouth, nausea, vomiting, fatigue, drowsiness, euphoria, confusion, hallucination, and loss of balance [67]. The long-term adverse effects of medical cannabis use are not known. One prospective cohort study followed 431 patients with chronic pain for one year and compared patients using medical cannabis with nonusers [75]. There was no difference in serious adverse events between the two groups. However, the medical cannabis group had a higher rate of non-serious respiratory adverse events. (See "Cannabis use: Epidemiology, pharmacology, comorbidities, and adverse effects".)

Botulinum toxin — Limited literature suggests that subcutaneous injection of botulinum toxin type A (BTX-A), a potent neurotoxin, may reduce opioid requirement for analgesia in patients with severe post herpetic neuralgia [76]. (See "Postherpetic neuralgia", section on 'Botulinum toxin'.)

Adjuvants we do not recommend

Muscle relaxants — We avoid using the drugs referred to as muscle relaxants (eg, methocarbamol, metaxalone, carisoprodol) for patients with chronic pain. A wide variety of pain conditions may be accompanied by painful muscles, and at times spasm. However, there is no evidence that these medications directly relax muscles. Muscle relaxants have diverse pharmacologic actions, but none of them act directly on muscle itself [77]. Pain relief and relief of spasm without spasticity is likely related to CNS effects, including sedation, rather than analgesic effects. When true muscular spasticity is present, anti-spasticity drugs, such as baclofen or tizanidine, may alleviate the pain from persistent tonic muscular contractions. (See "Chronic complications of spinal cord injury and disease", section on 'Spasticity'.)

Carisoprodol should no longer be used for any indication, due to lack of proven efficacy, high rates of physical dependence, and risk of agitation and delirium tremens when abruptly withdrawn.

Cyclobenzaprine is an alternative initial drug for patients with mild to moderate symptoms of fibromyalgia. It is a tricyclic closely resembling amitriptyline, with a presumed mechanism unrelated to muscle relaxation. Cyclobenzaprine is US Food and Drug Administration (FDA) approved for short-term use only. Doses and efficacy for patients with fibromyalgia are discussed separately. (See "Initial treatment of fibromyalgia in adults", section on 'Tricyclic antidepressants and related drugs'.)

Tizanidine and cyclobenzaprine are best studied for acute low back pain, with minimal evidence of efficacy for chronic back pain [78]. (See "Subacute and chronic low back pain: Nonpharmacologic and pharmacologic treatment", section on 'Pharmacologic therapy for persistent symptoms'.)

Muscle relaxants have anticholinergic properties, and may also cause CNS depression. They should be used cautiously in older patients, and when combined with other CNS depressant medications or antidepressants.

Benzodiazepines — We avoid the use of benzodiazepines in patients with chronic pain, including those with anxiety or post-traumatic stress disorder. Benzodiazepines are not first-line anxiolytics or sleep aids and there is no evidence of analgesic efficacy for chronic pain. Disadvantages include their abuse and addictive potential, and importantly, potentiation of respiratory depression and an increase in all-cause mortality with concomitant opioid use (see "Use of opioids in the management of chronic non-cancer pain", section on 'Drug interactions'). In a study of 1220 patients with non-cancer pain on long-term opioids, concurrent benzodiazepine use was associated with greater pain severity, prescription of higher doses of opioids, substance use, and greater mental health comorbidities [79].

OPIOIDS

General considerations — Based on compelling evidence of risk and absent strong evidence of effectiveness [80,81], long-term opioids should not routinely be used for chronic pain. If opioids are necessary, they should be used at the lowest effective dose, with ongoing reassessment of risks and benefits [82]. Similar concerns exist for children and adolescents, though available related evidence is scant and of low quality [83].

Opioids should only be used when the benefits outweigh the potential risks, and generally when other therapies have failed to provide adequate pain relief and improvement in function. Opioids should always be combined with nonpharmacologic and often nonopioid pharmacologic therapy, and be carefully monitored for benefit, risk, and treatment adherence.

Chronic opioid therapy should be reserved for patients who are assessed to be at low risk for substance abuse, who have persistent pain despite ongoing multimodal non-drug treatments, trials of nonopioid analgesics and antidepressants or antiepileptics, and in whom the potential benefits outweigh the risks. For patients with comorbidities that increase the risk of opioid related respiratory depression (eg, sleep apnea or other respiratory, significant renal, or hepatic insufficiency), opioids should be avoided if possible.

If opioids are indicated, immediate-release (IR) formulations should be used at the lowest effective dose, and naloxone should be co-prescribed. Extended-release (ER) or long-acting (LA) opioids should be reserved for only those patients who are anticipated to and informed that they will most likely remain continuously on opioids long term. Use of combination ER/LA opioids with IR opioids for "breakthrough pain" should be limited to those on ER/LA opioids and are receiving intermittent painful procedures as part of their care (eg, cancer-related pain). Use of IR opioids for pain flares is discouraged for patients on ER/LA opioids, since this contributes to dose escalations, and is better managed with non-pharmacological interventions (eg, cognitive behavioral therapy [CBT] and other mind-body strategies discussed earlier). Opioids used in combination with sedative drugs increase the risk of respiratory depression, and so should be avoided.

For patients who are already on long-term opioid analgesic therapy, consider tapering opioids to a lower dose or discontinuing entirely if long-term opioid analgesic therapy has not meaningfully improved pain and function, increasing doses have not been beneficial, opioid side-effects have reduced quality of life or function, or if benefit-harm balance is unclear. This approach is consistent with the 2018 United States Health and Human Services Guide for Clinicians on the Appropriate Dosage Reduction or Discontinuation of Long-Term Opioid Analgesics [84]. (See "Use of opioids in the management of chronic non-cancer pain", section on 'Discontinuing therapy' and "Opioid tapering for patients with chronic pain".)

While all patients should be periodically considered for opioid dose reduction, benefits of opioids may outweigh risks for those selected patients with persistent severe chronic non-cancer pain despite adherence to nondrug and nonopioid therapies (including compliance with primary disease modifying therapies). In the authors' experience, examples include persistent or recurrently severe nociceptive pain conditions (eg, inflammatory arthritis, inflammatory bowel disorders, sickle cell-related pain) and inflammatory or injury-related neuropathic pain (eg, autoimmune demyelinating disorders, spinal cord injury, and occasionally phantom pain and complex regional pain syndrome [CRPS]).

●(See "General principles and overview of management of rheumatoid arthritis in adults", section on 'Use of analgesics'.)

●(See "Approach to functional gastrointestinal symptoms in adults with inflammatory bowel disease", section on 'Therapies to avoid'.)

●(See "Overview of the management and prognosis of sickle cell disease", section on 'Pain management'.)

●(See "Chronic complications of spinal cord injury and disease", section on 'Pain syndromes'.)

●(See "Lower extremity amputation", section on 'Phantom limb pain'.)

The use of opioids for management of chronic non-cancer pain is discussed in more detail separately. (See "Use of opioids in the management of chronic non-cancer pain".)

Tramadol — Tramadol is a mixed mechanism opioid with a weak affinity for the mu opioid receptor and also serotonin and norepinephrine reuptake inhibition. Like other opioids, it may be used as second-line agent for patients with fibromyalgia who have not responded to initial therapy with other agents. Efficacy of tramadol for other types of chronic pain, including neuropathic pain, is unclear [4]. (See "Use of opioids in the management of chronic non-cancer pain", section on 'Choice of agent and dose' and "Treatment of fibromyalgia in adults not responsive to initial therapies", section on 'Analgesics'.)

Combination of tramadol with other serotonin reuptake inhibitors can increase the risk for serotonin syndrome/serotonin toxicity and seizures. For specific interactions, use the Lexicomp drug interactions program included with UpToDate.

Tapentadol — Tapentadol is also a mixed mechanism opioid with stronger affinity for the mu opioid receptor than tramadol (though weaker than pure mu agonists) and noradrenergic reuptake inhibition. It is the only opioid with a specific US Food and Drug Administration (FDA) indication for neuropathic pain (ER form for painful diabetic neuropathy). However, while tapentadol may be associated with fewer GI side effects than potent opioids at similarly effective doses, the authors reserve it for third-line use. Guidelines for treatment of neuropathic pain do not distinguish tapentadol from other opioids with respect to risk or efficacy. (See "Management of diabetic neuropathy", section on 'Not recommended'.)

Like tramadol, combination of tapentadol with other serotonin reuptake inhibitors can increase the risk for serotonin syndrome/serotonin toxicity and seizures may be increased. For specific interactions, use the Lexicomp drug interactions program included with UpToDate.

Buprenorphine — Buprenorphine is an opioid that is often used for maintenance pharmacotherapy for opioid use disorder (OUD), but it may also be prescribed for chronic pain. Buprenorphine may confer some safety advantages when prescribed for chronic pain, since it produces less physical dependence and less opioid-induced hyperalgesia than other opioids, and produces less respiratory depression than other LA opioids [85]. However, buprenorphine is associated with the same safety concerns as other opioids. Significant respiratory depression can occur if buprenorphine is used along with other benzodiazepines, alcohol, or other respiratory depressants. In addition, buprenorphine is an abusable drug.

For chronic pain indications, buprenorphine is available as a transdermal patch or a buccal film; these products are available in the United States in low-moderate dosage formulations, lower than the doses typically used for OUD. We and others [86,87] consider off-label use of the higher dose sublingual formulations for patients who are attempting to taper long-term full mu opioid therapy but are unable to do so. Use of buprenorphine in this setting has been endorsed by the US Health and Human Services. (See "Opioid tapering for patients with chronic pain".)

Use of buprenorphine for chronic pain is discussed separately. (See "Use of opioids in the management of chronic non-cancer pain", section on 'Buprenorphine for chronic pain'.)

INFUSION THERAPIES

Ketamine — Ketamine is an N-methyl-D-aspartate (NMDA) antagonist primarily used as a dissociative anesthetic agent. It has long been used in hospital and emergency settings to provide analgesia in acute pain situations (burns, fractures and other trauma, postoperative pain), but in recent years it has been increasingly used as an intravenous infusion for complex regional pain syndrome (CRPS), neuropathic pain, and other intractable chronic pain states. While there are no high-quality data to support ketamine infusions, the available data suggest that infusions given for several hours repeated over several days may have the potential to provide weeks to months of relief [88-90], with the best evidence of efficacy in CRPS patients. (See "Complex regional pain syndrome in adults: Treatment, prognosis, and prevention", section on 'Less favorable risk:benefit ratio'.)

Most studies of the efficacy of ketamine for chronic pain measure pain intensity as a primary outcome. Ketamine has effects beyond analgesia, and is used for treatment-resistant unipolar depression and suicidal ideation. Thus, it is possible that improved mood may be an important factor contributing to improved pain scores in patients who receive ketamine for chronic pain.

There are several reports of hepatobiliary dysfunction after recurrent ketamine administration [91-93]. In 2020 this prompted an FDA label change for ketamine, recommending that liver function tests should be checked before starting therapy, and periodically for those receiving recurrent therapy.  

Further study is required before recommending ketamine infusion for specific types of chronic pain, and for recommending doses, administration regimens, or adjuvant medications.

The use of ketamine infusion for treatment of depression is discussed separately. (See "Ketamine and esketamine for treating unipolar depression in adults: Administration, efficacy, and adverse effects".)

Lidocaine infusion — Lidocaine is an amide local anesthetic most commonly used for local or regional anesthesia, less commonly to treat ventricular arrhythmias. Intravenous lidocaine infusion is also used for multimodal analgesia for perioperative pain, and has been used to treat chronic neuropathic pain. In outpatient settings doses of 3 to 5 mg/kg administered over 30 to 60 minutes can relieve neuropathic pain short term, with sustained analgesia in some patients [94,95]. A positive response to IV lidocaine is sometimes viewed as an indication to proceed with other local anesthetics (typically mexiletine) or other sodium channel blocking agents, administered orally. Optimal doses, administration regimens, and patient selection, have not been determined. Lidocaine infusion should be reserved for patients who have not responded to other treatments.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic pain management".)

SUMMARY AND RECOMMENDATIONS

●Plan for treatment – Effective treatment of pain requires multimodal analgesia with an emphasis on non-drug modalities (eg, self-management, behavioral health support and physical therapy). When necessary, we add multi-targeted pharmacologic therapies (algorithm 1). (See "Approach to the management of chronic non-cancer pain in adults", section on 'Creating a plan for treatment'.)

●Type of pain – The choice of pharmacologic therapy depends on the type of chronic pain syndrome. In particular, neuropathic pain should be distinguished from nociceptive pain since treatments differ (table 1). (See 'General principles of drug selection' above.)

•Patients with nociceptive pain – For these patients the choice of pharmacologic therapy depends partly on the location of the pain, and also on patient comorbidities. Oral or topical nonsteroidal antiinflammatory drugs (NSAIDs) are the first-line therapy for many chronic nociceptive pain conditions. (table 6) If usual treatment is ineffective for patients who are thought to have predominantly nociceptive pain, it could be presumed that the patient has neuropathic or centralized pain and treatment should be changed. (See 'Pharmacologic therapy for nociceptive pain' above.)

•Patients with neuropathic pain – For these patients initial treatment involves antidepressants (ie, tricyclic antidepressants [TCAs], serotonin norepinephrine reuptake inhibitors [SNRIs]), or antiepileptic medications (gabapentin or pregabalin), with adjunctive topical therapy (eg, topical lidocaine, 8% capsaicin patch) when pain is localized (table 3). The choice among treatments should be based on the pain condition (if known), comorbidities, medication side effects, cost, and patient values and preferences. (table 5) (See 'Pharmacologic therapy for neuropathic pain' above.)

●Opioids – Opioids should be used on a chronic basis only in patients who are assessed to be at low risk for substance abuse, who have persistent pain despite trials of nonopioid analgesics and antidepressants or antiepileptics, and in whom the potential benefits outweigh the risks. Opioids should always be combined with nonpharmacologic and often nonopioid pharmacologic therapy, and be carefully monitored for benefit, risk, and treatment adherence. (See 'Opioids' above.)

●Antidepressants – Tricyclic antidepressants (TCAs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) are first line treatments for many chronic pain conditions, independent of their antidepressant effects. Selective serotonin reuptake inhibitors (SSRIs) are not first-line treatment for any chronic pain condition. Analgesic effects may require two to four weeks for maximal effect. These drugs have a variety of adverse effects which may limit their use. (table 7) (See 'Antidepressants' above.)

•TCAs – Amitriptyline, doxepin, imipramine, nortriptyline, and desipramine are used for chronic pain. Amitriptyline is the most sedating of these drugs. (See 'Tricyclic antidepressants' above.)

•SNRIs – Duloxetine, venlafaxine, and milnacipran are used for a variety of types of chronic pain. (See 'Serotonin norepinephrine reuptake inhibitors' above.)

●Antiepileptic medications – Antiepileptic medications are among the first line therapies for some forms of neuropathic pain.

•Gabapentinoids – Gabapentin and pregabalin are fist line therapies for painful diabetic neuropathy and postherpetic neuralgia. Of note, these drugs are associated with respiratory depression in older adults and in patients who receive other sedatives or opioids, and there is potential for abuse. (See 'Gabapentin and pregabalin' above.)

•Other antiepileptics – Carbamazepine is a first line treatment for trigeminal neuralgia. An alternative is oxcarbazepine. (See 'Other antiepileptics' above.)

●Adjuvant medications – Topical lidocaine or capsaicin and cannabinoids may be beneficial in some patients. We avoid the use of muscle relaxants (eg, tizanidine, cyclobenzaprine, carisoprodol) and benzodiazepines in patients with chronic pain. (See 'Other adjuvant medications' above.)

●Emerging therapies – Ketamine and lidocaine infusion are emerging therapies with mixed results for chronic pain. Optimal doses, administration regimens, and patient selection, have not been determined. (See 'Infusion therapies' above.)